CN103229535B - The method and system recovered for unit in communication network - Google Patents
The method and system recovered for unit in communication network Download PDFInfo
- Publication number
- CN103229535B CN103229535B CN201180055718.5A CN201180055718A CN103229535B CN 103229535 B CN103229535 B CN 103229535B CN 201180055718 A CN201180055718 A CN 201180055718A CN 103229535 B CN103229535 B CN 103229535B
- Authority
- CN
- China
- Prior art keywords
- unit
- resource
- fault
- core
- relevant
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/04—Arrangements for maintaining operational condition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W88/00—Devices specially adapted for wireless communication networks, e.g. terminals, base stations or access point devices
- H04W88/12—Access point controller devices
Abstract
Open a kind of method and system, assists in ensuring that and any unit collapses the individual unit that (accident behavior i.e. occurred as the result of software error or fault) is localized to support on the individual modem plate of multiple-unit configuration.Thus, chain of command and the residue unit configured in modem boards should keep operable.Additionally, operator should be able to select to use correct behavior (i.e. restart, reconfigure, delete or set up) for the unit in modem boards in the case of not affecting the operation of unit of other configurations.
Description
Background technology
The exemplary embodiment of the present invention relates to the list in the multiple-unit configuration supporting to use polycaryon processor
Unit recovery process in modem boards, wherein said polycaryon processor has and comprises all process
Single SMP partition of device core.Present invention is particularly directed to field of wireless telecommunications, therefore by special to it
Surely quote and describe, it is to be appreciated that the present invention is the most effective in other field and application.
As background technology, LTE(Long Term Evolution) it is intended to improve UMTS(universal mobile telecommunications
System) mobile phone standard with solve future communications network demand quick evolution in 3GPP plan.
LTE improves wireless network efficiency and bandwidth, reduces cost and promotes service experience.Specifically, LTE
Use new spectrum opportunities and the more preferable fusion with other open standards is provided.LTE generally includes LTE
RAN(radio access network) (being also known as E-UTRAN) and EPS(evolution packet system
System, also known as evolution block core).
Communication system generally falls into two major functions: data surface function and chain of command function.Previously
LTE product in, in modem boards use at least two processor: one support chain of command
Function (in non real-time, such as, operates, manages and manages (or OA&M), and call process
The function that management is relevant), another terminates and supports data surface function (in real time, such as, LTE
Layer 2 processes).Control and data surface all uses different operating system (OS) examples, such as, be used for
The Linux of chain of command and the real-time OS for data surface core, such as vXWorks(is by California
My rice wind reaches wind river system system (Wind River Systems of Alameda, California) and makes
And sell).Typically, a modem boards supports a sector or unit.So, in order to
Support multiple-unit (such as 3 unit or 6 unit) configuration, it is necessary to provide same with number of unit
The modem boards that sample is many.
As improvement, LTE wireless base station can use in modem boards polycaryon processor.
In the case of this, such as, there is the operating system of the SMP Linux of PREEMPT RT patch
Comprising a SMP(symmetric multi-processors of all 8 cores) on subregion.In this configuration, control
Identical operating system is shared with data surface (real-time thread/process) in face processed (non-real-time threads/process)
Example, even if they necessarily operate in different cores.
Owing to modem boards supports multiple unit and sector, so unit recovers to become complicated.
Previously in scheme (i.e. use a modem boards to support each unit), software crash it
After unit recovery process be simple.So, whole modem boards is restarted.Then, pass through
Suitable initialized all hardware and component software restart modem boards, are ready for use at controller board
The main OA&M-C of upper operation, to reconfigure unit in modem boards.
Restart modem boards with recover or rescue unit new architecture be no longer feasible option because
The residue active unit making chain of command and modem support is also restarted by this.High availability is to close
Service provider's demand of key, and must restart after software crash or again reconfigure only one
All unit not acceptable option is destroyed during individual unit.
Accordingly, it would be desirable to a kind of method and system, assist in ensuring that and the collapse of any unit (is lacked as software
The random operation that sunken or fault result occurs) it is localized to a unit.Thus, chain of command and
In modem boards, the residue unit of configuration should keep operability.Additionally, operator should be able to select
Select for the unit in modem boards take corrective action (i.e. restart, reconfigure, delete,
Or set up), and do not affect the operation of the unit of other configurations.
Summary of the invention
Described herein is for being intervened by operator or automatically in software fault (or collapse)
The most individually restart effective unit recovery process of (or recovery) discrete cell, and do not affect modulation
Other active units on demodulator plate.
In one embodiment, it is provided that a kind of computer implemented method, supporting that there is polycaryon processor
Multiple-unit configuration individual modem plate on provide unit Restoration Mechanism.Detection is derived from fault
One or more defects of unit, and to one or more component software notification units in fault.
Additionally, the resource that release is relevant to the unit in fault, and collect postmortem analysis data.And then,
Be provided for the execution environment of the new unit of trustship, and operation on remote controllers plate, operation,
Inform with management (OA&M) entity and can start new unit in modem boards.
In another embodiment, it is provided that a kind of system, for supporting to use the multiple-unit of polycaryon processor
Unit is provided to recover on the individual modem plate of configuration.This system includes: modem boards；
And polycaryon processor, including being connected to multiple processor cores of modem boards, wherein modulatedemodulate
Device plate is adjusted to include kernel spacing and user's space.This system also includes the list being configured to perform each function
Unit's Restoration Mechanism.These functions can include such as: unit one or more that detection is derived from fault
Defect；To one or more component software notification units in fault；Discharge and the unit phase in fault
The resource closed；Collect postmortem analysis data；It is provided for the execution environment of the new unit of trustship；With to far
Operation, operation and management (OA&M) entity on range controller plate are informed in modem boards
Upper energy starts new unit.
In another embodiment, the non-provisional computer data available carrier of storage instruction is being held by computer
Computer performance element recovery process is made during row.Unit recovery process can include each function.Such as,
Such function comprises the steps that the one or more defects detecting the unit being derived from fault；To one or
Multiple component software notification units are in fault；Discharge the resource relevant to the unit in fault；Collect
Postmortem analysis data；It is provided for the execution environment of the new unit of trustship；With on remote controllers plate
Operation, operation and management (OA&M) entity are informed and can be started new unit in modem boards.
Other scopes of the application of the present invention are made apparent from according to detailed description of the invention presented below.So
And, it is appreciated that the detailed description of the invention of instruction the preferred embodiment of the present invention is only given by diagram
And particular instance, because within the spirit and scope of the present invention each changes and amendment is for this area
Technical staff is made apparent from.
Accompanying drawing explanation
The present invention is stored in the structure of the various piece of equipment, deployment and combination, and the step of method
In Zhou, the most here as more completely illustrated, point out ground, Yi Ji the most in the claims
The purpose of design is obtained like that, in the accompanying drawings shown in accompanying drawing:
Fig. 1 illustrates an embodiment of platform architecture according to aspects of the present invention；
Fig. 2 illustrates the exemplary architecture realizing the core level of abstraction shown in Fig. 1；
Fig. 3 illustrates exemplary cell recovery process according to aspects of the present invention；With
Fig. 4 illustrates the example process of termination process.
Detailed description of the invention
Referring now to accompanying drawing, wherein this diagram is only used for illustrating exemplary embodiment, is not intended to limit
The theme advocated, Fig. 1 provides can be in conjunction with the view of the system of presently described embodiment.
Referring now to Fig. 1, it is shown that exemplary platform framework 100.This framework is usually used for modulation /demodulation
On device plate, it is to be appreciated that he can be used in other application.In this embodiment, wherein with complete
8, portion core defines a subregion.It can be appreciated, however, that polycaryon processor 100 can have arbitrarily
Number core.By this embodiment, therefore, it is possible to use in whole cores (such as 8 cores)
One symmetric multi-processors (SMP) operating system (OS) example 102 of upper operation.Due to chain of command
With data surface under an OS example, it usually needs ensure that problematic data surface no longer destroys control
Face.
In this example, polycaryon processor 100 serve 3 unit (be shown as 104 in the drawings,
106 and 108).Each unit need up-link (UL) scheduler (figure is shown as 110,
112 and 114) and downlink (DL) scheduler (being shown as 116,118 and 120 in Fig. 1).
Known use radio link controls (RLC) layer to be carried out segmentation, connect and revises across LTE empty
The mistake in packet that middle interface sends and receives.At GPRS(2.5G) wireless stack uses wireless
Current source road controls and media interviews control (RLC/MAC) software.There is provided and control at movement station and base station
The data transmission of acknowledgement and non-acknowledgement between device processed (BSC).Therefore, framework 100 also includes
RLC/MAC frame 122, this is the basis biography in the air interface used between the mobile station and the network
Defeated unit.RLC/MAC frame 122 is generally used for carrying data and RLC/MAC signaling.
Polycaryon processor 100 also provides for operation, operation and management (OA&M) module 124 and calling
Process (or CALLP) module 126.OA&M is generally used in description communication network operating, transporting
Battalion, manage and safeguard process that assembly relates to, behavior, instrument, standard etc..CALLP module 126
Typically manage the non real-time aspect of call treatment behavior.
Additionally, polycaryon processor 100 includes core level of abstraction (CAL) 128, it is generally directed to layer 2
(L2) application software hides the specific details of core.Layer 2 is seven layers of open systems of computer network
The data link layer of interconnection (OSI) model.Data link layer is adjacent networks joint in the wide area network
The protocol layer of data is transmitted between node between point or in same local network section.Data link layer carries
For transmitting function and the process apparatus of data between network entity, and detection can be provided and can revise
The device of the mistake that can occur within the physical layer.The example of SDL be LAN (multinode),
Peer-peer protocol (PPP), HDLC and ADCCP that connect for point-to-point (binode)
Ethernet.In the case of this, L2 be commonly referred to as LTE air interface needed for L2 scheduler
Processing, this can have the most intensive real-time requirement.
(it is responsible for processing mobile communication equipment and network exchange to meet the real-time performance demand of base station
Business between system and signaling), the SMP such as with PREEMPT RT patch can be used
The operating system of Linux.It is understood, of course, that, other operating systems can be used.In order to such
SMP configuration realizes definitiveness behavior, it is preferable that to use core to retain and core affinity structure
This system can be realized in the way of system action compared with asynchronous multiprocessing (AMP) by realization.This
It is also expected to obtain optimum performance from the SMP Linux such as with PREEMPT RT.Multiple without lock zero
The use of uniform business (such as buffer management and messenger service) can also contribute to solution can be by having
Any delay issue that the use of the SMP Linux of PREEMPT RT operating system causes.
One of major function of core level of abstraction (128) as shown in Figure 1 is for senior application (such as
L2 process) the various application of repertoire utilizing multi-core platform are provided.Therefore, by core level of abstraction
It is designed as realizing some purposes.First, he supports BED(backboard Ethernet driver) interface, this
Accelerate framework (DPAA) based on new data path, simultaneously for higher level application software (i.e.
L2 software) hide DPAA and the specific scheme of multinuclear.(it is designed as DPAA optimizing many nucleoreticulums
Network processes, and the load distribution of such as resource is with shared, including network interface and hardware accelerator.) its
Secondary, he utilizes the DPAA nextport hardware component NextPort of P4080, is user face number in entrance and exit both direction
According to providing the data path accelerated.3rd, he provides motility as much as possible, to be easily adaptable to
Configuration change (i.e., it is not necessary to code changes).The example of CAL configuration is for buffer unit pool
DPAA resource distribution, ingress frame queue and egress frame queue.
Commonly known system storage in the operating system of such as Linux can be divided into two not same districts
Territory: kernel spacing and user's space.Kernel spacing is that kernel (i.e. the core of operating system) performs (i.e.
Run) and its space serviced is provided.
Memorizer generally includes RAM(random access memory) unit, its content can be by high
Speed is accessed (i.e. read and write), but is only temporarily held (i.e., in use or at most,
When power supply is held open).Its object is to currently used in program and data.
User's space is depositing of running wherein of consumer process (any affairs outside the most unless the context)
The set of memory location.Process is the execution example of program.One of role of kernel is that management is at this
Each consumer process in space, and prevent them interfering with each other.
The use that kernel spacing can only be called by system is accessed by consumer process.System call be by
Request in the operating system of the similar Unix that active process is initiated, the service performed for kernel, example
As input/output (I/O) or process are set up.Active process is the current process of progress in CPU,
It is opposed in CPU, wait the process that its next time is run.I/O be to or from CPU and to or from
Ancillary equipment (such as disk drive, keyboard, mouse and printer) transmission data random procedure,
Operation or equipment.
Therefore, kernel spacing is strictly reserved, and is used for running kernel, kernel extensions and great majority
Device driver.On the contrary, user's space is the memory area of all user mode application work, and
And this memorizer can swap out when necessary.
Referring now to Fig. 2, it is shown that realize the exemplary architecture 200 of these and other purposes.Thus, core
Heart level of abstraction (CAL) 201 includes modules in the user space, includes but not limited to: core
Level of abstraction initializes (CALInit) module 202, it is configured LTE network and any static state resolves,
Classification and distribution (PCD) rule are loaded onto frame supervisor (FMan) 230 and 232, and based on joining
CAL framework is set up in the set putting file；Core level of abstraction buffer (CALBuf) module 204；
Core level of abstraction message (CALMsg) module 206, its to L2 software providing message service with to or
Send from another plate (i.e. eCCM) and receive user face data；Core level of abstraction resolve, classification and
Distribution (CALPcdFmc) 208, it provides by each FMan(230,232) PCD used
Rule and be configured to route to ingress frame suitable core；Follow the tracks of with core level of abstraction DPAA
(CALDpaaTrace) module 210, it provides following function for as kernel spacing module
Core level of abstraction DPAA driver (CALDpaaDriver) 212 in enable and disable tracking.
Framework 200 also includes suitable operating system 214, such as, have PREEMPT RT patch
SMP Linux.Subsequently, operating system 214 supports each driver, the most above-mentioned CALDPaa
Driver 212, at least one frame supervisor (FMan) driver 216, at least one buffer tubes
Reason device (BMan) driver 218 and at least one queue management device (QMan) driver 220.
As in figure 2 it is shown, framework 200 can suitably include that P4080CoreNet organizes 222, this is
It is suitable for expansible network-on-chip and multiple power architecture process core is connected to cache, unit height
Speed caching and the interconnection architecture of memory sub-system.
P4080 processor includes the realization of DPAA.Therefore, framework 200 may also include P4080
DPAA224.DPAA224 is designed to optimize multi-core network and processes, and such as, the load of resource divides
Cloth is with shared, including network interface and hardware accelerator.As it can be seen, DPAA224 generally distinguishes
Each including such as BMan226, QMan228 and the first and second Fman230 and 232
Individual manager.
It is known that in wireless many access communication systems, emitter and receptor can use multilamellar communication stack
Communication.Described layer can include that such as physical layer, media interviews control (MAC) layer, radio link
Control (RLC) layer, protocol layer (such as packet data convergence protocol (PDCP) layer), application
Layer etc..Rlc layer receives service data unit (SDU) from PDCP layer, and connects or by SDU
It is segmented into rlc protocol data cell (PDU), for being transmitted to MAC layer.
Thus, CALBuf module 204 promote for RLC SDU process in for L2 application
Without lock buffer management service.As it is known in the art, non blocking algorithm guarantees the line for shared resource
Cheng Jingzheng will not make their execution by mutual exclusion indefinite duration postpone.If secured system model
The process enclosed, then non blocking algorithm is without lock (or exempting from lock).CALBuf module 204 also may be used
Support inquiry (such as, pond consumption state, consumption notation, the Chi Ke for buffer unit pool statistical data
With sexual state, pond assignment error counting etc.).CALBuf module 204 is generally and CALDpaaDriver
212 docking are to realize such service.CALBuf module 204 also provides for without lock buffer management side
Case, this operates for the suitable system in multi-core environment is extreme crucial, wherein by Non real-time processing
The lock used can cause delay issue for the processing in real time of release waiting this lock.
CALDpaaDriver212 is the kernel spacing assembly of CAL201, and use Bman and
Qman API helps to realize and provide buffer management service and messenger service.Here, term API
(or application programming interface) refers to the interface realized by software program so that it is can be with other software
Alternately.It is mutual that he promotes between different software procedures, be similar to user promote user and computer it
Between mutual mode.API is realized by application, storehouse and operating system, to determine theirs
Vocabulary and calling convention, and it is used for accessing their service.Can include the specification of routine, data structure,
Object class and be used between consumer and the implementer of API communication agreement.
CALDpaaDriver212 be generally responsible for managing DPAA resource to be used (buffer unit pool and
Frame queue) it is distributed for user face data；Via each file operation (such as open, close,
I-o controls (ioctl)) provide user's space interface for initializing, delaying to other CAL modules
Rush device management and messenger service；Perform kernel to map to user's space (K-U) buffer；There is provided
DPAA buffer pool and receptor and emitter statistical data；And it is slow to realize being used for managing annular
Rush the service of device.It should be noted that the L2 software queue that circular buffer generally represents CAL, and he
Be used for store the frame descriptor (FD) pointing to specific L2 Downlink scheduler thread.CALMsg
Module 206 provides API for L2, to extract buffer descriptor from ring.
All CAL assemblies described above are usually platform middleware (running in user's space),
Except CALDpaaDriver212.CALDpaaDriver212 be in kernel spacing run determine
Driver processed, and its be designed to realize and provide by the clothes needed for CAL user's space middleware
Business-especially, depend on those services of P4080DPAA nextport hardware component NextPort.
CALBuf module 204 provides and is exclusively used in the buffer management clothes that " fast path " data process
Business.CALBuf module 204 provides user's space API to L2 application.CALBuf module 204 with
CALDpaaDriver212 cooperates, and sets up but by Bman226 for CALDpaaDriver212
The buffer of management provides zero duplication and without lock buffer management service.
As it has been described above, for the performance requirement meeting modem boards based on polycaryon processor, nothing
Lock buffer management scheme is important, and its use includes a subregion of whole 8 cores and passes through
PREEMPT RT runs SMP Linux.Not having such scheme, system can be prolonged through unlimited
Spike late, this can damage whole system.
Now, we forward the feelings of in fault or the fault of a unit in modem boards to
Shape.About exemplary embodiment, as broad as long between unit and the failed unit in fault.
In the case of any one, final result is to discharge the resource that such unit obtains.Thus, can realize
Unit recovery process.One of primary demand for unit recovery process is that the recovery of unit should shadow
Ring the function of other unit in the identical modem boards of unit in trustship fault, behavior and
Performance.Additionally, unit recovers not introduce other work can having a strong impact in modem boards
The unstability of moving cell or shortage of resources.
Turning now to Fig. 3, exemplary cell recovery process is shown in flow charts.Unit set forth below
The basic function of recovery process:
Detection is derived from fault one or more defects of the unit of (or failed), and to by shadow
The component software rung sends suitable message (310)；
The resource (320) that release (or removing) is relevant to the unit in fault；
Collect the postmortem analysis data (330) relevant to the unit in fault；With
Foundation can the execution environment of the new unit of trustship the OA&M-C on remote controllers plate
Notice can start new unit (340) in modem boards.
It is understood that exemplary cell recovery process is frequently not individual software component.On the contrary, unit is extensive
Multiple process suitably includes be bound in other component softwares of the many relevant to modem boards one
Group integrating process.Unit recovery process applies various rule about all component softwares of administrative unit resource
Then.Specifically, modulation and demodulation software assembly should follow each common design and coding rule.
First, unit recovery process should be able to participate in or detect under they are responsible for any other are soft
Part and/or the fault of nextport hardware component NextPort.Mistake can be predicted by periodicity extraction state and statistics.Equally,
Unit recovery process should be in predetermined time period detection fault (can use software and hardware house dog).
The health of any run entity should be monitored.Additionally, for unit distribution or activation system resource the softest
Part assembly is necessarily these resources and provides reset procedure.
Suitably, framework 200 also includes " trigger " process, and this is to initiate all software processes (bag
Include layer 2 scheduler handle) middleware component.It should be noted that enode b (eNodeB) is supported
LTE air interface and perform in fault/provided for radio resources management behavior that failed unit is relevant.
Thus, LTE eNodeB application is one group of cooperating process.These process need by individually, can
The process leaned on starts and monitors.This supervisory process is referred to as trigger.Trigger is unit and data
Drive, receive via a group profile unlatching, monitor and recover the data needed for application process.
Trigger is started by Linux init process (via opening script), and subsequent start-up constitutes application
Every other process.By design, trigger the most only uses Linux to service.Trigger process
It it is the parent process of all application processes.Trigger generally " is intercepted " regularly and is terminated event (bag with process
Include layer 2 process to terminate) relevant information.
When the code fault of layer 2 process relevant to unit, (that is, the layer 2 not controlled by code enters
Terminating unintentionally of journey), before it terminates, still there are some behaviors occurred in the environment of process.
The environment of the Linux the most typically thread in fault is initiated signal handle, and exits by LEC
The handle that (Linux error collection device) and other storehouses are set up.Noting, some exits handle and is registered in
LEC is the most called with gather information when process terminates subsequently.Linux uses signal handle to send
Signalisation.Linux calls " closedown " of each driver opened by the process in fault pellucidly
Function.Specifically, he calls fast path driver CALDpaaDriver212.This is at kernel
The chance of clearing cell related resource under pattern.If be configured at layer 2 process termination, then Linux
Generate Core Dump.
Fig. 4 illustrates the sequence list of process termination phase.Originally, operating system 102 starts (for answering
With code transparent ground) error collection device handle, to collect the data (410) for error collection snapshot.
Operating system 214 subsequent start-up (for application code pellucidly) code tracking handle, to collect generation
Code tracking daily record (420).Then, operating system 214 is called and is currently opened by the process in fault
" closedown " function (430) of each kernel driver module, to perform by kernel driver module
Any cost for the course allocation in fault is removed.Finally, operating system 214 is to " trigger "
Notify about the process of termination (440).Alternatively, trigger also can receive termination event, including terminating
State.Then, trigger is notice and the resource reset procedure that system scope is triggered in failed community.
Polycaryon processor also supports " process monitors ", and this is also middleware component.Process monitors
" issuing/customization " service is used to issue state and the availability of monitored program.Such as, may be used
Via Linux message queue, process monitors is connected to trigger.Process monitors is that trigger is entered
The subprocess of journey.Here, term " subprocess " refers to be set up by another process (parent process)
Process.Each process can set up many subprocess, but will only have a parent process, except first
Process, it does not have parent process.First process (being referred to as init in Linux) is being drawn by kernel
Time of leading opens, and never terminates.
When trigger receives process termination event (such as SIGCHLD Linux signal), he
This event is sent to process monitors via message queue.Here, SIGCHLD refers to enter at son
The signal sent to process when journey terminates.Using issue/customization procedure, process monitors is issued layer 2 and is entered
The final state of journey, as " state changes instruction " for this process.
When it initializes, the local OA&M module 124 in modem boards customizes such as to enter
Layer 2 process " state changes instruction " that journey monitor is issued.In order to customize, OA&M module 124
Should know the identity of all layer 2 processes, they are typically predefined form.Work as Message-based IPC
When layer 2 process " state changes instruction " issued by device, OA&M module 124 receive termination message.
When OA&M module 124 receives the instruction about the termination of layer 2 process, (excellent from shared memorizer
Selection of land is realized by kernel module) extract the letter of reason about fault (such as dsp software collapse)
Breath.OA&M module 124 runs on remote controllers plate to OA&M-C() send message,
To stop the every other eNodeB behavior relevant to failed unit as early as possible.
If still the assembly in fault distributes such element resources after the fault of component software
Proprietary rights, then the reset procedure that must activate correspondence is correlated with to relinquish possession and to discharge corresponding unit
Resource.This demand mainly processes the application process allocation unit resource representing request in kernel spacing
P4080 driver.At cell location device, self is registered in by L2 process in the user space
CALMsg module 206.Subsequently, CALMsg module 206 is via CALDpaaDriver212
L2 course allocation resource (kernel, hardware) for request.
In modem boards, the most fatal and expendable software or hardware fault can to service
Have an impact by property.As it was earlier mentioned, modem boards generally includes the many shared by multiple unit
Resource.Therefore, the purpose of exemplary cell recovery process generally includes: closes related software and cancels
Element resources, and do not affect operating unit；The unit can being again turned in fault and optimization clothes
Business availability；And collect the purpose that the enough data about fault diagnose for postmortem analysis.
Owing to the software fault in the range of unit recovery process is corresponding to relevant to unit whole soft
Losing of part assembly, so the telecommunication environment of correspondence can not be recovered by unit recovery process.Unit
Recovery process cannot avoid the service disruption for the unit in fault inevitably, but its main purpose
For minimizing this lasting event interrupted and scope.
It is important to note that, the arbitrarily collapse relating to chain of command process or thread the most typically affects tune
All unit of configuration on modulator-demodulator plate.Therefore, fair termination is processed, is sent out by operator
Go out unit reset command, L2 process or thread should realize before terminating self, perform its unit all money
The logic of " removing " (or the release) in source.
Management in a number of ways is the P4080 system resource of unit distribution.Such as, some resources are from being
The shared pool distribution of system scope.When distributing such resource (i.e. being had) by unit, they cannot
By other component reuses, until they are clearly discharged.At the software having the resource from shared pool
In the case of fault, platform software must represent owner's software and remove these resources.Otherwise, corresponding
Resource will forever lose.In order to remove these resources, platform needs to identify which unit has them
Mode.To this end, each personal resource from shared pool has the possessory relevant of appointment resource
Label.Owner's software should update when each allocation unit resource and every time when shared pool discharges.
" fast path " is to have the more brachydactylia by program compared with " normal route " for description
Make the term in the path of path.For effective fast path, it is necessary to more more effective than normal route
Ground solves the task that great majority occur, and makes the latter process uncommon situation, and individual cases, at mistake
Reason, and other are abnormal.Fast path is the form optimized.The entrance that fast path driver uses delays
Rushing device pond is the example that system scope shares resource.They are exclusively used in reception user plane packets.
It should be noted that and distribute some other resources from privately owned pond.Resource from privately owned pond need not be with holding
The owner of resource carrys out labelling.This is because, when distributing such resource, impliedly know all
Person.Platform software only needs to know the state of unit；That is, unit whether resource is busy or resource is idle.
It should be noted that and set up element resources when arranging execution environment, but need not Resources allocation.When configuration is patrolled
Distribution is made when collecting unit.The output port buffer pond that fast path driver uses is that unit is the most private
There is the example of resource.They are exclusively used in transmission user plane packets.Should be recovered by unit as quickly as possible
Journey discharges resource that is that distribute and that had by the unit fault from shared pool.
When layer 2 process surprisingly terminates, its resource can be completed in the period in one of following two stage
Removing, i.e. (1) detect the termination of process once operating system, or (2) are when as at the beginning of unit
Beginningization and/or configuration a part produce layer 2 scheduler process new example time after.Without spy
Determining problem, the removing for the element resources of layer 2 scheduler process should preferably occur in the layer of correspondence
After 2 scheduler process terminate.
Can be for following reason clearing cell resource: (1) concedes space to pass through for residue unit
Enough system resource is run；(2) if setting up the new example of unit, resource and memorizer are avoided
Consumption.
Some stage allocation unit software resources in " life-span " of unit.Distribute when dispensing unit
Some software resources.According to the behavior of unit, distribute continually and discharge some other resources.One
Such example is communication buffer.In the case of Gai, about entering packet, kernel to user be empty
Between (K-U) map buffer and obtained by suitable frame supervisor, and must be by the unit in fault
Downlink scheduler discharges.When the software fault of processing unit, some communication buffers still by
Component software in fault has.
Usually, the specific resource of all unit (such as, buffer, context, data structure,
Interruption, thread, suspension DMA etc.) can be affected by cell failure.But, the side of management resource
Formula is specific to application architecture, and the behavior of suitable operating system (such as Linux).Such as, exist
Between user's space resource and kernel spacing resource and the privately owned pond of resource and resource shared pool it
Between can be distinct.Similarly, from central entities (such as CALLP module 126 or OA&M
Module 124) distribution compared to from unit special entity (such as layer 2 scheduler handle) point
Can be distinct between joining.
The memorizer management some virtual storage addressings of management subsystem of the operating system of such as Linux
Space a: address space is for each process, and an address space is for kernel.When process terminates
Time, operating system is automatically releasable all resources from process address space distribution.Subsequently, it is not necessary to make
The clearly removing of resource for a part for the process address space.Platform software driver (example should be passed through
Such as CALDpaaDriver212) remove the list that the unit in representing fault distributes from kernel address space
Unit's specific resources.It should be noted that the termination of process is the most automatically from kernel module clearing cell resource.
Term used herein " is removed " and is not only represented the storage buffer in kernel spacing, and
It is also represented by other kernel resources many and the behavior specific to unit.Such kernel money is below described
Some examples of source and behavior.
By example, if the result as unit collapse deletes the unit managed, then can need to prohibit
Interrupt with some.
Equally, if (such as the result of unit collapse) deletes the unit managed, then can need
Disable some DMA transfer.DMA represents direct memory access, and can represent in calculating system
Equipment in system or other entities revise the ability of main-memory content in the case of without CPU.
Driver can use kernel thread to be used for housekeeping, statistics or other cycle behaviors.If example
Result as collapsed as unit deletes the unit managed, then can disable these behaviors.
According to fast path, the configuration of DPAA and Linux requirement definition, first by data surface
Entry data stream is assigned to corresponding P4080 core, is then assigned to be matched with Downlink scheduler line
The correct of the buffer descriptor of journey enters choma.Specifically, for data stream fast path and
DPAA configures and should at least meet following demand:
1. the logic IP addresses of definition unit in modem boards；
2. the frame queue relevant to the logic IP addresses of unit should be tied to the core phase with trustship unit
The QMan entrance closed；With
3. the context of the frame queue relevant to unit should point to the buffer relevant with object element and describe
Accord with enters choma.
Recover about unit, the previous configuration of the entry data stream for unit should be cancelled.In order to cancel
This configuration, is considered as some behaviors.Such as, it is called when CALLP module 126 and notifies to close
In the information of cell failure, then should be from the logic IP addresses of modem boards removal unit.Additionally,
When notifying the information that " fast path " CALMsg module 206 terminates about layer 2 scheduler process,
Mapping between ingress frame queue can be changed back to chain of command core from current data face core.
It should be noted that the CALMsg module 206 of Fig. 2 includes each API, comprise registration and cancellation is stepped on
Cite sb. for meritorious service energy.Thus, the commonly provided some processes, for representing the element resources of layer 2 course allocation, are wrapped
Include but be not limited to: (1) CALMsg registration function is (for each data stream type: entrance and going out
Mouthful)；(2) CALMsg cancels registration function (for each data stream type: entrance and exit)；
(3) driver " closes " process (it covers all data stream type).
The CALMsg registration function request fast path kernel module of BED interface is outlet and entrance
Resource needed for data stream type foundation.CALMsg registration function also asks fast path kernel module
Surplus resources is removed from the previous case of identical layer 2 scheduler process.Layer 2 scheduler process from
CALLP module 126 receives the CALMsg registration merit calling BED interface when unit arranges request
Energy.
It is for by layer by layer 2 scheduler process from the output port buffer that privately owned output port buffer pond is distributed
The possible candidate that next example of 2 processes is purged.These resources correspond to outlet data stream type.
One reason of the removing postponing these resources provides time transmission and release to suitable frame supervisor
The demand of all packets (being queued up by L2 uplink scheduler before it collapses).
The output port buffer of buffer-manager is a part for privately owned buffer unit pool, and resets buffer unit pool
It it is relatively simple operation.Therefore, message and need not be is set when layer 2 scheduler process receives unit
Set up layer 2 scheduler process constantly, carry out reseting buffer unit pool.
Except the point of penetration corresponding for CAL API specified by BED interface section, such as
The fast path driver of CALDpaaDriver212 must provide " closedown " point of penetration.By applying
Any driver called should provide open and close point of penetration.It is different from by CAL storehouse
Other points of penetration exposed, are not exposed to layer 2 for " closedown " point of penetration of fast path module and adjust
Degree device application.Layer 2 scheduler application should not directly invoke " closedown " for fast path driver
Function.
" closedown " function for driver is commonly used for the standard Linux function of any driver.
If process terminates, then Linux finds the All Drives opened by the process terminated and for all
These drivers " close " point of penetration one by one." close " function and can determine that calling L2 scheduler enters
The PID(process ID of journey) and the core number that is cited." closing " function can be by requesting party
PID and corresponding unit ID associates." close " function and can remove the list being had by corresponding unit ID
The ring of unit resource, such as buffer descriptor.
OA&M module 124 can be from FFS(flash memory file system) some core dump file in
Determine which file is corresponding to the final state received from process monitors.So, he can manage and be used for
The memory space of Core Dump is for postmortem analysis.Postmortem analysis data generally include Core Dump and
Code tracking daily record, it can be used for the root cause analysis of unit collapse.OA&M module 124 is attempted really
Guarantor complete resource in need remove (such as quote CALBuf API with release fault in unit
Independent entry data buffer in using.Execution environment is set with can the new unit of trustship, i.e.
Must arrange can before carrying out cell location by OA&M-C new unit need all softwares and
Hardware resource.Therefore, the local OA&M module 124 in modem boards is to the most main
OA&M-C notice can configure new unit on this plate again.
Restart modem boards to need a few minutes and affect all unit of plate trustship.The most single
Unit recovers only need several seconds and the most only affect the unit in fault.Generally, collect for dividing afterwards
The data of analysis are the longest steps of unit recovery process.The step for typically can not shorten because number
Word signal processor (DSP) cannot be restarted during its memory dump.Other unit recovering step
Duration need not to be critical, because they are at TTI(Transmission Time Interval) labelling (allusion quotation
Type ground was less than 1 second) middle measurement.
For being exclusively used in the component software of unit, there are the various possibilities realizing unit recovery process.One
Individual mode is the component software (relevant to the unit in fault) only stopped in fault and removes relevant
Resource with make other assemblies (relevant to the unit in fault) and be rolled back to original state and do not stop him
?.Because the strong interdependency between the relative broad range of likelihood of failure and component software, this side
Case is complicated and is not without a breath of air danger.Preferably scheme can be to stop with " strongly "
Assembly, and restart the component software of all participations relevant to the unit in fault subsequently, even if only one
Individual (suspection) breaks down.
It can be easily understood by those skilled in the art that the step of each said method can pass through programmed computer
Perform.Here, some embodiments also aim to overlay program storage device, such as, machine or calculating
The digital data storage medium that machine is readable, and the machine of coded command can perform or computer can perform
Program, wherein said instruction performs the some or all of steps of approach described above.Program storage device
Can be such as digital memory, magnetic storage medium (such as Disk and tape), hard disk drive,
Or the most readable digital data storage medium.Embodiment also aims to covering and is programmed to execute above-mentioned side
The computer of the described step of method.
Above description only provides the disclosure of only certain embodiments of the present invention, the purpose being not intended to limit.
Thus, the present invention is not limited only to above-described embodiment.On the contrary, it can be appreciated that, those skilled in the art can
It is susceptible to the alternative fallen within the scope of the present invention.
Claims (19)
1. a system, provides multiple-unit configuration in the modem boards in communication network
Unit recovers, and this system includes:
Modem boards；With
Polycaryon processor, it is connected to described modem boards, and wherein said polycaryon processor includes
Multiple processor cores, the plurality of processor core configure under single symmetric multi-processors subregion and by
Thering is provided the single operating example service of Compliance control face and N number of data surface, wherein N is described tune
Configured element number on modulator-demodulator plate, wherein said polycaryon processor is configured at least perform
Following functions:
One or more defects of the unit that detection is derived from fault；
Identify the unit in described fault and distribute to described fault from the system scope shared pool of resource
In the system resource of unit；
The resource of the unit that release is assigned in described fault；
Perform in described fault in the case of not affecting described Compliance control face or other operating unit
The recovery of unit；
The execution environment of the new unit of trustship it is provided for by following steps:
The logic IP addresses of the unit in described fault in configuration and described modem boards
Relevant resource；
The frame queue relevant to the logic IP addresses of described new unit is tied to described in trustship
The queue management device entrance that the core of new unit is correlated with is to set up packet transmission path；
The context of the described frame queue relevant to described new unit is pointed to and divides with incoming data surface
The buffer descriptor that the object element of group is relevant enter choma；With
Operation, operation and management OA&M entity on remote controllers plate are informed at described tune
New unit can be started on modulator-demodulator plate.
2. the system as claimed in claim 1, wherein unit recovers do not restarting or reseting described multinuclear
Processor or realize in the case of restarting the single operating example that described control and data surface are shared.
3. system as claimed in claim 2, wherein said polycaryon processor is detecting cell failure
Time be further configured to:
Notice call processing module is to extract the cellular logic IP ground removed from described modem boards
Location；
Mapping between one or more ingress frame queues is changed back to chain of command from current data face core
Core, in order to still process the incoming packet with particular core as target, until joining on the core
Put new unit；
Disable the interruption of elementary layer 2 process in described fault；
Disable the direct memory access transmission relevant to the unit in described fault.
4. system as claimed in claim 2, wherein said polycaryon processor is detecting layer 2 process
It is further configured to during termination:
Notice call processing module is to extract the cellular logic IP ground removed from described modem boards
Location；
Mapping between one or more ingress frame queues is changed back to chain of command from current data face core
Core, in order to still process the incoming packet with particular core as target, until joining on the core
Put new unit；
Disable the interruption of elementary layer 2 process in described fault；
Disable the direct memory access transmission relevant to the unit in described fault.
5. the system as described in claim 3 or 4, wherein said polycaryon processor is further configured
For discharging allocated resource from the system scope shared pool of described resource by following steps, so that
The unit described shared pool of return in described fault:
Perform extra statistics, in order to each personal resource in the system scope shared pool of communication buffer
There is the possessory respective labels specifying described resource；
Update institute when in each allocation unit resource and discharging this element resources from described shared pool every time
State owner's label.
6. the system as claimed in claim 1, wherein said polycaryon processor clear and definite calling platform software
Driver, in order to when described operating system is only automatically releasable from the resource of process address space distribution,
Release represents the unit specific resources that the unit in described fault distributes from kernel address space.
7. the system as claimed in claim 1, wherein exists to one or more component software notification units
In fault, described component software includes the OA&M entity in described modem boards and described tune
At least one of call processing module on modulator-demodulator plate.
8. the system as claimed in claim 1, wherein said polycaryon processor is further configured to lead to
Cross following steps and collect post-flight data to perform basic reason analysis:
Start multiple error collection device handle, to collect the data for error collection snapshot；
Start multiple code tracking handle, to collect multiple code tracking daily record.
9. the system as claimed in claim 1, wherein said polycaryon processor is further configured to lead to
Cross following steps and collect post-flight data to perform basic reason analysis:
Start multiple error collection device handle, to collect the data for Core Dump；
Start multiple code tracking handle, to collect multiple code tracking daily record.
10. the system as claimed in claim 1, wherein said polycaryon processor is further configured to
Stop the component software relevant to the unit in described fault, then restart and the unit in described fault
Relevant described component software.
11. 1 kinds of computer implemented methods, carry in the modem boards in communication network
Unit for multiple-unit configuration recovers, and the method includes:
Cause by the unit in fault by being connected to the polycaryon processor detection of described modem boards
One or more defects, wherein said polycaryon processor includes multiple processor core, the plurality of
Processor core configures and by providing Compliance control face and N number of data under single symmetric multi-processors subregion
The single operating example service in face, wherein N is the configured unit in described modem boards
Quantity；
Identify the unit in described fault and distribute to described fault from the system scope shared pool of resource
In the system resource of unit；
The described resource of the unit that release is assigned in described fault；
Perform in described fault in the case of not affecting described Compliance control face or other operating unit
The recovery of unit；
The execution environment of the new unit of trustship it is provided for by following steps:
The logic IP addresses of the unit in described fault in configuration and described modem boards
Relevant resource；
The frame queue relevant to the logic IP addresses of described new unit is tied to described in trustship
The queue management device entrance that the core of new unit is correlated with is to set up packet transmission path；
The context of the described frame queue relevant to described new unit is pointed to and incoming data bread
The relevant buffer descriptor of object element enter choma；And
Operation, operation and management OA&M entity on remote controllers plate are informed at described tune
New unit can be started on modulator-demodulator plate.
12. methods as claimed in claim 11, wherein unit recovers do not restarting or reseting described many
Core processor or realize in the case of restarting the single operating example that described control and data surface are shared.
13. methods as claimed in claim 12, farther include:
When cell failure being detected, notice call processing module is to retrieve from described modem boards
The cellular logic IP address removed；
Mapping between one or more ingress frame queues is changed back to chain of command from current data face core
Core, in order to still process the incoming packet with particular core as target, until joining on the core
Put new unit；
Disable the interruption of elementary layer 2 process in described fault；
Disable the direct memory access transmission relevant to the unit in described fault.
14. methods as claimed in claim 12, farther include:
When detecting that layer 2 process terminates, notice call processing module is to retrieve from described modulation /demodulation
The cellular logic IP address that device plate is removed；
Mapping between one or more ingress frame queues is changed back to chain of command from current data face core
Core, in order to still process the incoming packet with particular core as target, until joining on the core
Put new unit；
Disable the interruption of elementary layer 2 process in described fault；
Disable the direct memory access transmission relevant to the unit in described fault.
15. methods as claimed in claim 11, farther include by following steps from described resource
System scope shared pool in discharge allocated resource, so that unit in described fault returns described
Shared pool:
Perform extra statistics, in order to each personal resource in the system scope shared pool of communication buffer
There is the possessory respective labels specifying described resource；
Update institute when in each allocation unit resource and discharging this element resources from described shared pool every time
State owner's label.
16. methods as claimed in claim 11, the clear and definite calling platform of wherein said polycaryon processor is soft
Part driver, in order to when described operating system is only automatically releasable from the resource of process address space distribution,
Release represents the unit specific resources that the unit in described fault distributes from kernel address space.
17. methods as claimed in claim 11, farther include to lead to one or more component softwares
Know unit in fault, wherein said one or more described component softwares include described modem
At least one of OA&M entity on plate and the call processing module in described modem boards.
18. methods as claimed in claim 11, farther include to be collected by following steps to count afterwards
Execution basic reason analysis according to this:
Start multiple error collection device handle, to collect the data for error collection snapshot；
Start multiple code tracking handle, to collect multiple code tracking daily record.
19. methods as claimed in claim 11, farther include to be collected by following steps to count afterwards
Execution basic reason analysis according to this:
Start multiple error collection device handle, to collect the data for Core Dump；
Start multiple code tracking handle, to collect multiple code tracking daily record.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US12/950,096 | 2010-11-19 | ||
| US12/950,096 US8730790B2 (en) | 2010-11-19 | 2010-11-19 | Method and system for cell recovery in telecommunication networks |
| PCT/US2011/058726 WO2012067810A1 (en) | 2010-11-19 | 2011-11-01 | A method and system for cell recovery in telecommunication networks |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103229535A CN103229535A (en) | 2013-07-31 |
| CN103229535B true CN103229535B (en) | 2016-10-19 |
Family
ID=45003058
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201180055718.5A Expired - Fee Related CN103229535B (en) | 2010-11-19 | 2011-11-01 | The method and system recovered for unit in communication network |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US8730790B2 (en) |
| EP (1) | EP2641419B1 (en) |
| JP (1) | JP5710017B2 (en) |
| KR (1) | KR101619078B1 (en) |
| CN (1) | CN103229535B (en) |
| WO (1) | WO2012067810A1 (en) |
Families Citing this family (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8634302B2 (en) * | 2010-07-30 | 2014-01-21 | Alcatel Lucent | Apparatus for multi-cell support in a network |
| US8504744B2 (en) | 2010-10-28 | 2013-08-06 | Alcatel Lucent | Lock-less buffer management scheme for telecommunication network applications |
| US8737417B2 (en) | 2010-11-12 | 2014-05-27 | Alcatel Lucent | Lock-less and zero copy messaging scheme for telecommunication network applications |
| US8861434B2 (en) | 2010-11-29 | 2014-10-14 | Alcatel Lucent | Method and system for improved multi-cell support on a single modem board |
| US9357482B2 (en) | 2011-07-13 | 2016-05-31 | Alcatel Lucent | Method and system for dynamic power control for base stations |
| CN103813366B (en) * | 2012-11-07 | 2017-12-08 | 锐迪科（重庆）微电子科技有限公司 | A kind of abnormal restoration methods of processor and mobile terminal |
| CN103839015A (en) * | 2013-03-18 | 2014-06-04 | 株式会社Ｇｅｎｕｓｉｏｎ | Storage device |
| JP2014206967A (en) * | 2013-03-18 | 2014-10-30 | 株式会社Ｇｅｎｕｓｉｏｎ | Storage device |
| CN106063313B (en) * | 2013-10-10 | 2019-10-01 | 诺基亚通信公司 | Cancel the change made to communication network |
| CN103945456B (en) * | 2014-05-12 | 2017-06-27 | 武汉邮电科学研究院 | A kind of efficient UDP message of LTE base station user plane based on linux system sends optimization method |
| CN106293905B (en) * | 2015-06-09 | 2019-03-22 | 宇龙计算机通信科技(深圳)有限公司 | Resource regulating method and device based on multi-core processor |
| CN107733681B (en) | 2017-07-28 | 2018-10-30 | 贵州白山云科技有限公司 | A kind of scheduling scheme configuration method and device |
| CN108154431B (en) * | 2018-01-17 | 2021-07-06 | 北京网信云服信息科技有限公司 | Target recruitment state processing method and device |
| US11025547B2 (en) * | 2019-01-15 | 2021-06-01 | Vmware, Inc. | Enhanced network stack |
| CN111800810B (en) * | 2020-06-30 | 2022-09-13 | 展讯通信（上海）有限公司 | Intelligent device and system and method for recovering abnormity of WCN module thereof |
| CN113760448A (en) * | 2021-04-30 | 2021-12-07 | 中科天玑数据科技股份有限公司 | Big data management platform based on kubernets |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1788491A2 (en) * | 2005-11-16 | 2007-05-23 | Alcatel Lucent | Thread aware distributed software system for a multi-processor array |
| CN101207519A (en) * | 2007-12-13 | 2008-06-25 | 上海华为技术有限公司 | Version server, operation maintenance unit and method for restoring failure |
| CN201243299Y (en) * | 2008-07-02 | 2009-05-20 | 广东天波信息技术有限公司 | Asymmetric digital loop modem system |
Family Cites Families (93)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5295133A (en) * | 1992-02-12 | 1994-03-15 | Sprint International Communications Corp. | System administration in a flat distributed packet switch architecture |
| US6141346A (en) | 1995-07-19 | 2000-10-31 | Fujitsu Network Communications, Inc. | Point-to-multipoint transmission using subqueues |
| US5913230A (en) | 1997-01-07 | 1999-06-15 | Richardson; John J. | Object and method for providing efficient multi-user access to shared operating system kernal code using instancing |
| US6434620B1 (en) | 1998-08-27 | 2002-08-13 | Alacritech, Inc. | TCP/IP offload network interface device |
| US6314501B1 (en) | 1998-07-23 | 2001-11-06 | Unisys Corporation | Computer system and method for operating multiple operating systems in different partitions of the computer system and for allowing the different partitions to communicate with one another through shared memory |
| US7299052B2 (en) * | 1998-11-09 | 2007-11-20 | Samsung Electronics Co., Ltd. | System and method for reallocating a traffic channel in soft handoff state |
| US6606298B1 (en) * | 1999-06-30 | 2003-08-12 | Cisco Technology, Inc. | Module firmware recovery |
| US6842811B2 (en) | 2000-02-24 | 2005-01-11 | Pts Corporation | Methods and apparatus for scalable array processor interrupt detection and response |
| US6999432B2 (en) | 2000-07-13 | 2006-02-14 | Microsoft Corporation | Channel and quality of service adaptation for multimedia over wireless networks |
| US6799200B1 (en) | 2000-07-18 | 2004-09-28 | International Business Machines Corporaiton | Mechanisms for efficient message passing with copy avoidance in a distributed system |
| US6584330B1 (en) | 2000-07-18 | 2003-06-24 | Telefonaktiebolaget Lm Ericsson (Publ) | Adaptive power management for a node of a cellular telecommunications network |
| US6735620B1 (en) | 2000-07-18 | 2004-05-11 | International Business Machines Corporation | Efficient protocol for retransmit logic in reliable zero copy message transport |
| US7089289B1 (en) | 2000-07-18 | 2006-08-08 | International Business Machines Corporation | Mechanisms for efficient message passing with copy avoidance in a distributed system using advanced network devices |
| JP2002050996A (en) | 2000-07-31 | 2002-02-15 | Sony Corp | Communication system transmitting signals coded using block lengths comprising with integral multiple interrelation via communication transmission path |
| US7089335B2 (en) | 2000-10-30 | 2006-08-08 | Microsoft Corporation | Bridging multiple network segments and exposing the multiple network segments as a single network to a higher level networking software on a bridging computing device |
| US7096034B2 (en) | 2001-10-01 | 2006-08-22 | Microsoft Corporation | System and method for reducing power consumption for wireless communications by mobile devices |
| US7768522B2 (en) | 2002-01-08 | 2010-08-03 | Apple Inc. | Virtualization of graphics resources and thread blocking |
| US7007042B2 (en) * | 2002-03-28 | 2006-02-28 | Hewlett-Packard Development Company, L.P. | System and method for automatic site failover in a storage area network |
| US7254812B1 (en) | 2002-05-31 | 2007-08-07 | Advanced Micro Devices, Inc. | Multi-processor task scheduling |
| WO2004001615A1 (en) | 2002-06-19 | 2003-12-31 | Telefonaktiebolaget Lm Ericsson | A network device driver architecture |
| US7093013B1 (en) * | 2002-06-19 | 2006-08-15 | Alcatel | High availability system for network elements |
| US7180866B1 (en) * | 2002-07-11 | 2007-02-20 | Nortel Networks Limited | Rerouting in connection-oriented communication networks and communication systems |
| US7020484B2 (en) | 2002-10-29 | 2006-03-28 | Qualcomm Incorporated | Controlling multiple modems in a wireless terminal using energy-per-bit determinations |
| US7539853B2 (en) | 2002-11-18 | 2009-05-26 | Arm Limited | Handling interrupts in data processing of data in which only a portion of a function has been processed |
| US7810124B2 (en) | 2003-01-28 | 2010-10-05 | Thomson Licensing | Robust mode staggercasting fast channel change |
| US7003597B2 (en) | 2003-07-09 | 2006-02-21 | International Business Machines Corporation | Dynamic reallocation of data stored in buffers based on packet size |
| ES2325957T3 (en) * | 2003-08-25 | 2009-09-25 | Interdigital Technology Corporation | RUNNING LINK OPERATION IMPROVED IN A SOFT TRANSFER. |
| EP1678920B1 (en) | 2003-10-15 | 2007-04-18 | NTT DoCoMo, Inc. | Apparatus and method for controlling an operation of a plurality of communication layers in a layered communication scenario |
| WO2005041533A1 (en) | 2003-10-15 | 2005-05-06 | Ntt Docomo, Inc. | Apparatus and method for controlling an operation of a plurality of communication layers |
| US7206966B2 (en) | 2003-10-22 | 2007-04-17 | Hewlett-Packard Development Company, L.P. | Fault-tolerant multi-core microprocessing |
| JP2005190038A (en) * | 2003-12-25 | 2005-07-14 | Hitachi Ltd | Diagnostic processing method and diagnostic processing program for processor |
| US7370326B2 (en) | 2004-04-02 | 2008-05-06 | Emulex Design & Manufacturing Corporation | Prerequisite-based scheduler |
| WO2005104576A1 (en) * | 2004-04-22 | 2005-11-03 | Utstarcom Telecom Co., Ltd. | A distributed wireless system for controlling the resource centrally |
| JP4414305B2 (en) | 2004-08-19 | 2010-02-10 | 富士通株式会社 | Relay device, relay device restart method, and relay device software program |
| US8888600B2 (en) | 2004-08-25 | 2014-11-18 | Igt | Emulation methods and devices for a gaming machine |
| US7352693B2 (en) * | 2005-01-31 | 2008-04-01 | Nextel Communications Inc. | Fault tolerant wireless communication systems and methods |
| US7933197B2 (en) * | 2005-02-22 | 2011-04-26 | Cisco Technology, Inc. | Method and apparatus for constructing a repair path around a non-available component in a data communications network |
| US7620753B1 (en) | 2005-03-17 | 2009-11-17 | Apple Inc. | Lockless access to a ring buffer |
| US8644246B2 (en) | 2005-07-05 | 2014-02-04 | Nokia Corporation | Scheduling information at serving cell change |
| GB0519981D0 (en) | 2005-09-30 | 2005-11-09 | Ignios Ltd | Scheduling in a multicore architecture |
| US7472261B2 (en) | 2005-11-08 | 2008-12-30 | International Business Machines Corporation | Method for performing externally assisted calls in a heterogeneous processing complex |
| DE602005007620D1 (en) | 2005-12-14 | 2008-07-31 | Ntt Docomo Inc | Device and method for determining the transmission policy for several and varied applications |
| US8072879B2 (en) * | 2006-02-03 | 2011-12-06 | Cisco Technology, Inc. | Technique for determining whether to reestablish fast rerouted primary tunnels based on backup tunnel path quality feedback |
| US8689025B2 (en) | 2006-02-21 | 2014-04-01 | Qualcomm Incorporated | Reduced terminal power consumption via use of active hold state |
| US7295952B1 (en) * | 2006-05-24 | 2007-11-13 | International Business Machines Corporation | Enabling high availability and load balancing for JMX Mbeans |
| US20080002702A1 (en) | 2006-06-30 | 2008-01-03 | Symbol Technologies, Inc. | Systems and methods for processing data packets using a multi-core abstraction layer (MCAL) |
| WO2008005793A2 (en) | 2006-06-30 | 2008-01-10 | Symbol Technologies, Inc. | Systems and methods for processing data packets using a multi-core abstraction layer (mcal) |
| US20080002681A1 (en) | 2006-06-30 | 2008-01-03 | Symbol Technologies, Inc. | Network wireless/RFID switch architecture for multi-core hardware platforms using a multi-core abstraction layer (MCAL) |
| CN101106490B (en) | 2006-07-11 | 2012-01-04 | 华为技术有限公司 | Establishment method of preset stream classifier and its system and user terminal |
| CN101491059B (en) | 2006-07-20 | 2013-02-13 | 高通股份有限公司 | Methods, apparatuses and computer readable media for sharing a network connection between a utility engine including GPS ad an application |
| US7873964B2 (en) | 2006-10-30 | 2011-01-18 | Liquid Computing Corporation | Kernel functions for inter-processor communications in high performance multi-processor systems |
| US8059532B2 (en) | 2007-06-21 | 2011-11-15 | Packeteer, Inc. | Data and control plane architecture including server-side triggered flow policy mechanism |
| US8279865B2 (en) | 2007-04-20 | 2012-10-02 | John Giacomoni | Efficient pipeline parallelism using frame shared memory |
| US8055822B2 (en) | 2007-08-21 | 2011-11-08 | International Business Machines Corporation | Multicore processor having storage for core-specific operational data |
| US8194699B2 (en) * | 2007-09-21 | 2012-06-05 | Intel Corporation | Radio scheduler and data plane interface |
| US20100296428A1 (en) | 2007-09-28 | 2010-11-25 | Pin-Han Ho | A robust system and method for wireless data multicasting using superposition modulation |
| US8527622B2 (en) * | 2007-10-12 | 2013-09-03 | Sap Ag | Fault tolerance framework for networks of nodes |
| EP2220884B1 (en) * | 2007-12-17 | 2019-08-14 | Telefonaktiebolaget LM Ericsson (publ) | Method and apparatus for mitigating cell outage |
| US7944815B2 (en) * | 2008-02-14 | 2011-05-17 | Allied Telesis Holdings K.K. | System and method for network recovery from multiple link failures |
| US7831710B2 (en) * | 2008-02-25 | 2010-11-09 | International Business Machines Corporation | Communication of offline status between computer systems |
| FI20085217A0 (en) * | 2008-03-07 | 2008-03-07 | Nokia Corp | Data Processing device |
| CN101546276B (en) | 2008-03-26 | 2012-12-19 | 国际商业机器公司 | Method for achieving interrupt scheduling under multi-core environment and multi-core processor |
| US8024417B2 (en) | 2008-06-04 | 2011-09-20 | Microsoft Corporation | Simple flow control protocol over RDMA |
| US20100029266A1 (en) | 2008-07-02 | 2010-02-04 | Nokia Corporation | System and methods for quality of experience reporting |
| US7957297B2 (en) * | 2008-07-14 | 2011-06-07 | Cisco Technology, Inc. | Termination message for wireless wide area network routers |
| EP2313830B1 (en) | 2008-08-13 | 2013-10-02 | Hewlett-Packard Development Company, L.P. | Dynamic utilization of power-down modes in multi-core memory modules |
| US20100080116A1 (en) * | 2008-09-29 | 2010-04-01 | Qualcomm Incorporated | Re-establishing a radio resource control connection with a non-prepared base station |
| US8271996B1 (en) | 2008-09-29 | 2012-09-18 | Emc Corporation | Event queues |
| US8812714B2 (en) | 2008-11-12 | 2014-08-19 | Citrix Systems, Inc. | Systems and methods for application fluency policies |
| KR101546780B1 (en) | 2008-12-18 | 2015-08-25 | 삼성전자주식회사 | Apparatus and method for handling error for service flow modification in a broadband wireless communication network |
| WO2010084688A1 (en) * | 2009-01-20 | 2010-07-29 | シャープ株式会社 | Mobile station apparatus, base station apparatus, method for determining wireless link synchronization |
| JP5042248B2 (en) | 2009-01-22 | 2012-10-03 | 株式会社日立製作所 | Mobile communication system, call control server, and access gateway apparatus |
| CN101801015B (en) * | 2009-02-06 | 2014-03-12 | 中兴通讯股份有限公司 | Method and device for processing out of service faults of cell |
| KR101617048B1 (en) * | 2009-06-01 | 2016-05-02 | 엘지전자 주식회사 | Apparatus and method of processing failure of component carrier in multiple carrier system |
| US8099546B2 (en) | 2009-06-09 | 2012-01-17 | Red Hat, Inc. | Mechanism for a lockless ring buffer in overwrite mode |
| US8413153B2 (en) | 2009-06-12 | 2013-04-02 | Freescale Semiconductor Inc. | Methods and systems for sharing common job information |
| US20100322067A1 (en) * | 2009-06-19 | 2010-12-23 | Qualcomm Incorporated | Method and apparatus to faciliate reestablishing communications in a wireless network |
| US8737407B2 (en) * | 2009-06-22 | 2014-05-27 | Citrix Systems, Inc. | Systems and methods for distributed hash table in multi-core system |
| US9775046B2 (en) | 2009-06-26 | 2017-09-26 | Qualcomm, Incorporated | Power management |
| US8400955B2 (en) | 2009-09-21 | 2013-03-19 | Samsung Electronics Co., Ltd. | System and method for power saving by coordinated wake-up in a wireless multi-band network |
| US8892931B2 (en) | 2009-10-20 | 2014-11-18 | Empire Technology Development Llc | Power channel monitor for a multicore processor |
| US8737262B2 (en) | 2009-11-24 | 2014-05-27 | Red Hat Israel, Ltd. | Zero copy transmission with raw packets |
| US20130061231A1 (en) | 2010-05-11 | 2013-03-07 | Dong-Qing Zhang | Configurable computing architecture |
| US9298768B2 (en) | 2010-07-21 | 2016-03-29 | Sqream Technologies Ltd | System and method for the parallel execution of database queries over CPUs and multi core processors |
| US8634302B2 (en) * | 2010-07-30 | 2014-01-21 | Alcatel Lucent | Apparatus for multi-cell support in a network |
| US8943334B2 (en) | 2010-09-23 | 2015-01-27 | Intel Corporation | Providing per core voltage and frequency control |
| US20120093047A1 (en) | 2010-10-14 | 2012-04-19 | Alcatel-Lucent USA Inc. via the Electronic Patent Assignment System (EPAS) | Core abstraction layer for telecommunication network applications |
| US8504744B2 (en) | 2010-10-28 | 2013-08-06 | Alcatel Lucent | Lock-less buffer management scheme for telecommunication network applications |
| US8737417B2 (en) | 2010-11-12 | 2014-05-27 | Alcatel Lucent | Lock-less and zero copy messaging scheme for telecommunication network applications |
| US8861434B2 (en) | 2010-11-29 | 2014-10-14 | Alcatel Lucent | Method and system for improved multi-cell support on a single modem board |
| US8675577B2 (en) | 2010-12-20 | 2014-03-18 | Intel Corporation | Signaling techniques for a multimedia-aware radio and network adaptation |
| US20120207011A1 (en) * | 2011-02-11 | 2012-08-16 | Renesas Mobile Corporation | Method and apparatus for responding to an unrecoverable error |
| US9357482B2 (en) | 2011-07-13 | 2016-05-31 | Alcatel Lucent | Method and system for dynamic power control for base stations |
-
2010
- 2010-11-19 US US12/950,096 patent/US8730790B2/en not_active Expired - Fee Related
-
2011
- 2011-11-01 WO PCT/US2011/058726 patent/WO2012067810A1/en active Application Filing
- 2011-11-01 CN CN201180055718.5A patent/CN103229535B/en not_active Expired - Fee Related
- 2011-11-01 JP JP2013539862A patent/JP5710017B2/en not_active Expired - Fee Related
- 2011-11-01 KR KR1020137015368A patent/KR101619078B1/en active IP Right Grant
- 2011-11-01 EP EP11785842.3A patent/EP2641419B1/en not_active Not-in-force
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1788491A2 (en) * | 2005-11-16 | 2007-05-23 | Alcatel Lucent | Thread aware distributed software system for a multi-processor array |
| CN101207519A (en) * | 2007-12-13 | 2008-06-25 | 上海华为技术有限公司 | Version server, operation maintenance unit and method for restoring failure |
| CN201243299Y (en) * | 2008-07-02 | 2009-05-20 | 广东天波信息技术有限公司 | Asymmetric digital loop modem system |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2012067810A1 (en) | 2012-05-24 |
| EP2641419A1 (en) | 2013-09-25 |
| JP5710017B2 (en) | 2015-04-30 |
| US20120131376A1 (en) | 2012-05-24 |
| CN103229535A (en) | 2013-07-31 |
| JP2014504466A (en) | 2014-02-20 |
| KR20130086371A (en) | 2013-08-01 |
| KR101619078B1 (en) | 2016-05-10 |
| EP2641419B1 (en) | 2018-09-19 |
| US8730790B2 (en) | 2014-05-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103229535B (en) | The method and system recovered for unit in communication network | |
| CN103154897B (en) | Core level of abstraction for communication network application | |
| US10965558B2 (en) | Method and system for effective data collection, aggregation, and analysis in distributed heterogeneous communication network | |
| US6665262B1 (en) | Distributed fault management architecture | |
| US20180063762A1 (en) | Active hot standby redundancy for broadband wireless network | |
| US20220304114A1 (en) | Public safety analytics gateway | |
| JP3640187B2 (en) | Fault processing method for multiprocessor system, multiprocessor system and node | |
| EP1424808A1 (en) | Hierarchical management system of the distributed network management platform | |
| CN101502144A (en) | Element management system in wireless communication network | |
| US9098341B2 (en) | Service manager for adaptive load shedding | |
| JP6546340B2 (en) | Automatic symptom data collection in cloud deployment | |
| CN106453124A (en) | Traffic scheduling method and device | |
| US20220166664A1 (en) | Method and Apparatus for Obtaining Management Data | |
| CN103677988B (en) | The multi-process means of communication and system for software systems | |
| Rath et al. | MAQ system development in mobile ad-hoc networks using mobile agents | |
| JP2010502089A (en) | Component management system for wireless communication network with expandable structure | |
| CN103986709B (en) | Middleware registration method under RESTful framework | |
| CN104272789B (en) | GSM, call treatment node and communication control method | |
| CN100407619C (en) | Method for implementing switching between main and standby port by network processor | |
| CN113055416B (en) | Management method and device for container cluster node resource pool | |
| CN104657240B (en) | The Failure Control method and device of more kernel operating systems | |
| Li et al. | A Petri Net-based model to study the impact of traffic changes on 5G network resilience | |
| WO2023058137A1 (en) | Action execution system and method for controlling same | |
| WO2023058133A1 (en) | Action execution system and control method thereof | |
| US20230284053A1 (en) | Integration of physical test environments with a cloud-native cellular core |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20161019 Termination date: 20211101 |